Annotate a screenshot, free, in your browser
Arrows, boxes, circles, numbered pins, and text labels on any screenshot. Nothing to install, no signup to start, and the markup is burned into the image so it survives wherever you send it.
Annotating a screenshot is the fastest way to make a visual point: an arrow at the broken button beats a paragraph describing where the broken button is. This page covers doing it in a browser tab, free, with no extension and no signup. The rest of the solutions hub covers the specific tools this replaces.
What you can put on the image
Four shapes and a pin, which is the set that does almost all the work in practice:
- Arrows. Drag from the tail to the head. Point at the thing.
- Boxes and circles. Drag from corner to corner. Hold Shift to constrain to a square or a circle.
- Text labels. Click to place, type, press Enter. Single line, three sizes, positioned as a fraction of the image width so they scale with it.
- Numbered pins. Drop a marker and it takes the next number automatically. Delete one and the rest renumber themselves.
Everything comes in a seven-colour palette chosen to stay visible on real interfaces: red, orange, yellow, green, pink, black, white. Text labels get a contrasting halo drawn behind them, so light text stays legible on a light screenshot and dark text on a dark one, which is the failure mode most annotation tools leave you to solve by hand.
Drag a shape's outline to move it. Click its ✕ to delete it. Double-click a label to edit the words (on a phone, one tap).
The part that matters: the markup is in the file
Plenty of tools show you annotations in their own viewer and then hand you a bare screenshot the moment you export. CobaltCapture composes the marks into the PNG itself, on the server. The arrow you drew is in the pixels.
That means the marked-up image is what shows up in the markdown export, the PDF, the Word document, and the public share link, and it stays correct when someone drags the image into Slack or a ticket. It also means an AI coding agent reading the review sees the same arrow you drew, not a description of it.
Annotating a website, honestly
If you searched for a way to annotate a website, be clear on which of two things you want.
Some tools install a browser extension that pins comments onto the live page, so the markup is attached to the DOM and moves as the site changes. That's a real category, and if you need it, Markup.io and Pastel are built for it.
CobaltCapture works on a capture of the site: you grab the screen, mark up the still, and share it. You lose the live-DOM attachment. You gain that there is nothing to install, nothing to sign up for, and whoever opens your link needs neither — which is usually what decides whether a client actually replies.
Where a screenshot annotation is the whole deliverable
- Design review. Circle the spacing that's wrong, arrow the wrong font weight, send the link. See design review for the full workflow.
- Bug reports. A box around the broken state and a pin numbered
1:that your written repro steps refer to. - Client feedback. Mark up the build, add the question you want answered next to each mark.
- Quick how-tos. Numbered pins are a step list that lives on the image instead of underneath it.
What it does not do
Worth saying plainly, because the wrong tool wastes your afternoon:
- No multi-line or wrapped text callouts. Labels are one line each.
- No blur or redaction tool. If you need to hide a customer's data before sharing, crop it out.
- No scrolling or full-page capture. You annotate the frame you captured, not a stitched full-page image.
- No live-DOM pinning, as above.
If you need a full annotation suite with a shape library and OCR, Snagit and CleanShot X are genuinely better at that job, and both pages say so. This is the right tool when the annotated screenshot needs to reach someone else, fast, without either of you installing anything.
Frequently asked questions
How do I annotate a screenshot without installing anything?
Open CobaltCapture in a browser tab, add your screenshot (capture the screen, paste from the clipboard, drag the file in, or pick it with Add image), then choose Annotate and drag on the image to draw. Arrows, boxes, and circles are drag-to-draw; text labels are click-to-place. Nothing is installed and no account is needed to start.
Is there a free screenshot annotation tool?
This one. Capture, annotation, export, and public sharing are all free with no trial countdown and no per-seat plan. Most desktop annotation tools charge a one-time license or a subscription; the browser-based ones usually want a Chrome extension first.
What can I draw on the screenshot?
Arrows, rectangles, ellipses, and single-line text labels, each in a seven-colour palette, plus numbered pins for referencing a spot in your written notes. Hold Shift while dragging to constrain a box or circle to a perfect square or circle. Up to 24 marks per image.
Does the markup stay on the image when I share or download it?
Yes. The annotations are drawn into the PNG on the server, not layered over it in the browser, so the marked-up image is what appears in the markdown, PDF, and Word exports and what anyone sees at the share link. Nothing is lost when the file leaves the tool.
Can I annotate a website?
You annotate a capture of the website rather than the live page itself. Capture the screen (or paste a screenshot), mark it up, and share it. There is no browser extension pinning comments onto a live DOM, which is the tradeoff for needing no install and no signup from whoever opens the link.
Can I annotate a screenshot on my phone?
Yes. Take a screenshot the normal way, open the site, and use Add image, then draw with your finger. Tap a text label to edit it. Screen capture itself is desktop-only because mobile browsers have no API for it, but everything after that works on touch.
Capture your first review.
About a minute from open tab to a shareable URL your agent can ingest.
Start capturing